UNi HARVARD UNIVERSITY Library of the Museum of Comparative Zoology OCCASIONAL PAPERS ^ of the ^D MUSEUM OF NATURAL HISTORY The University of Kansas Lawrence, Kansas NUMBER 35, PAGES 1-51 APRIL 16, 1975 A REVIEW OF THE ANDEAN LEPTODACTYLID FROG GENUS PHRYNOPUS By John D. Lynch^ In 1932 Hampton Parker described Etipsophus wettsteini from the Peruvian Andes. In discussing the systematic position of this species he pointed out the then ominous possibiHty that Eleuthero- dactyltis, Etipsoplnis, and Sijnlwphus might eventually be merged into a single genus. Parker envisioned a "Borborocoetes" group of Eleutherodactijlus as having given rise to the annectant SyrrJwpus simonsii, S. festae, and Eupsophiis icettsteini, and that this complex of species had given rise to the more southern Etipsoplnis (espe- cially E. penianus). Although he separated the species into two genera, Parker evidently viewed them as an annectant generic group. Subsequently, these three species and several others were gathered into the genus Niceforonia (Lynch 1971). 1971 ten of in the Lynch ( ) recognized genera leptodactylids Telmatobiinae . Four of the trilpe Eleutherodactylinii ( ) genera {Eleutherodactijlus, Sminthilhis, Syrrliophus, and Tomodoctyhis) were characterized, in part, by the possession of digital discs and the associated T-shaped terminal phalanges. The other six genera {Amhlyphrynus, Euparkerella, Holoaden, Hylactophryne, Ischnoc- nema and Niceforonia) do not have digital discs and with the ex- ception of Euparkerella (which has complex terminal phalanges) have knobbed terminal phalanges. Only three eleutherodactyline genera have species represented in the Andean frog fauna at elevations above 2000 meters. The genus Amhlyphrynus is represented by a single species in Andean 1 Associate Professor, School of Life Sciences, The University of Nebraska, Lincoln, Nebraska; Research Associate in Herpetology, Museum of Natural History, The University of Kansas. 2 OCCASIONAL PAPERS MUSEUM OF NATURAL HISTORY this heads head width 50-60 Colombia. Frogs of genus have broad ( % SVL) and simple digits (no pads or discs; terminal phalanges knobbed). The genus Eleutherodactijlus is represented by many species in the Colombian and Ecuadorian Andes and few species elsewhere. These frogs have heads of normal width (HW 30-42% SVL) and complex digits (discs on narrowly to broadly dilated pads; terminal phalanges T-shaped ) . A third supraspecific group is represented by at least two Co- lombian, three Ecuadorian, eight Peruvian, and one Bolivian species. These frogs have heads of normal breadth and simple digits; in- cluded among the fourteen species are the type-species of the nominal genera Niceforonia Coin and Cochran, 1963, Nohlella Bar- bour, 1930, and Phrynopiis Peters, 1874. I consider the fourteen species congeneric and a distinct generic entity; accordingly, Nice- foronia and Nohlella are here relegated to the synonymy of Phryno- pus. In recent years, Niceforonia has been applied to some of the nominate species of Phrynoptis. Since 1932, Phrynopus penianus Peters has been considered a species of Eupsophus (Parker, 1932) and since 1971 the type-species of Nohlella has been considered a species of Eleutherodactylus (Lynch, 1971). Phrynoptis penianus and Niceforonia nana so closely resemble one another that their congeneric status cannot be seriously questioned. The third type- species (Nohlella peruviana) is more distinctive and is treated separately below. Presently, ten names are applied to the species of Phrynopus (nine are listed as Niceforonia by Lynch, 1971). One of the names {N. festae Peracca) was incorrectly applied to the species of Phryn- opus found in the upper Papallacta valley in Napo Province, Ecua- dor by Lynch (1968). Considerable material of this genus has accumulated since my 1968 paper; that material requires the recognition of four previously unknown species and enables me to review the genus as an entity and to discuss the variation in osteological character-states seen among the species of the genus. Two hundred and eighty-two specimens including 17 cleared and stained skeletons were studied. ACKNOWLEDGMENTS For loan of species or provision of working space at their respec- tive institutions, I am indebted to Jose M. Cei, the late Doris M. Cochran, William E. Duellman, Josef Eiselt, Alice C. C. Grandison, Arnold G. Kluge, the late James A. Peters, Douglas Rossman, Richard Thomas, Charles F. Walker, Ernest E. Williams, George Zug, and Richard G. Zweifel. Field work in Ecuador was supported by grants from the Com- mittee on Systematic and Evolutionary Biology, The University of Kansas, Sigma Xi, and the Watkins Fund of the Museum of Natural ANDEAN LEPTODACTYLID FROG GENUS PHRYNOPUS 3 History, The University of Kansas. Travel to museums was sup- ported by the University of Kansas Graduate School during my 1966-69 tenure as an Honors Fellow ( ) and The University of Nebraska Research Council. Robert Henderson and Marsha Lynch aided me in the field and William E. Duellman and Thomas H. Fritts provided detailed field notes and additional information about localities and the organisms they had collected. Todd Georgi aided in histological study of the pectoral girdle. The aid of all these persons has substantially contributed to my study of these frogs. Abbreviations for collections used throughout the text are: AMNH American Museum of Natural History British BMNH Museum ( Natural History ) IBM/UNC Instituto de Biologia, Mendoza-Universidad Nacional de Cuyo KU The University of Kansas Museum of Natural History LSUMZ Louisiana State University Museum of Zoology MCZ Museum of Comparative Zoology MNW Naturhistorisches Museum zu Wien UMMZ University of Michigan Museum of Zoology USNM United States National Museum ( National Museum of Natural History) WCAB Private collection of Werner C. A. Bokermann, Sao Paulo. STRUCTURE OF THE DIGITS Eleutherodactylus is a large and diverse genus; against this back- ground of morphological diversity, Fhnjnopus can be distinguished, at present, only on the basis of the structure of the digits. The majority of Eleutherodactylus species have dilated digital pads (bearing discs on their ventral surface) supported internally by distinctly T-shaped terminal phalanges. The disc is a distinct structure on the ventral surface of the digit and is defined by a circumferential 1 . In dila- groove ( Fig. ) most species having apical tion of the digit, the pad is wider than long and the dilation ratio (pad width/digit width below pad) is 2.0-3.5. This value is lowest when the digits bear lateral fringes (as in E. riveti). Several Eleu- therodactylus species have narrow digits, that is, the pad is longer than wide or the length and width are equal; the dilation ratio is as low as 1.2-1.3. The narrow-toed Eleutherodactylus include those with narrow finger and narrow toe pads {e.g., E. blnotatus, E. cor- nutus, E. elassodiscus, E. octavioi, and several others) as well as those E. anoma- with narrow finger pads and broad toe pads ( e.g., lus). In most of the narrow-toed species, the terminal phalanges of the fingers are more knobbed (as contrasted to T-shaped) whereas those of the toes are more T-shaped (Lynch, 1971). The 4 OCCASIONAL PAPERS MUSEUM OF NATURAL HISTORY most proximal digits (I and sometimes II of the hand; I and II of the foot) may not bear discs {i.e., no circumferential groove pres- ent) but the more distal digits always bear discs, even in the most narrow-toed species (e.g., E. elassodiscus, E. nigrovittatus, and 1 . E. octavioi; Fig. ) The simple-toed Andean frogs of the genus Fhrynopus do not have digital discs (Fig. 1). There is apical swelling (r= bulbing) of the digit in some of these frogs and dilation ratios of 1.0-1.2. Earlier (1971), I characterized this genus (as Niceforonia) as having knobbed terminal phalanges. However, some of the Peruvian taxa have lateral processes on the terminal phalanges (Fig. 2) that are intermediate in size between the T-shaped terminal phalanges of some of the narrow-toed Eleutherodactylus and the knobbed char- acter-state seen in Fhrynopus peraccai, P. favomaciihitu.s, and P. montium. in The annectant condition seen some Peruvian species ( no discs, Fig. 1.—Dorsal views of distal portion of (A) fourth toe, left foot, Fhrynopus peruvianus, (B) fourth toe, left foot, Eleutherodactyhis elassodiscus, (C) fourth finger, left hand, E. riveti. Ventral views of same digits, (A') P. peru- vianus, (B') E. elassodiscus, (C) E. riveti. Lateral views of same digits, (A") P. peruvianus, (B") E. elassodiscus, (C") E. riveti. Fhrynopus peru- vianus, KU 138959; Eleutherodactylus elassodiscus, WCAB 37915; E. riveti, KU 120060. Lines below lateral views equal 1 mm. ANDEAN LEPTODACTYLID FROG GENUS PHRYNOPUS 5 rudimentary or vestigal T-shaped terminal phalanges) could be used as evidence for combining several of the simple-toed genera of elutherodactylines with Elentherodactyhis as once noted by Parker (1932). To do so, howexer, does not simplify or improve leptodactylid taxonomy. Although the phalangeal characteristic is no longer diagnostic, the digital disc characteristic remains diag- nostic; accordingly, I prefer to retain the simple-toed genera as distinct from Eleutlierodactyliis. THE STATUS OF NOBLELLA BARBOUR Noble (1921) named SminthUhis penwianus on the basis of five specimens from Juliaca, Puno, Peru. The species was designated the type-species of NohJeUa by Barbour (1930), who neither diagnosed the of a nor defined genus. Lynch ( 1971), following study partially macerated, cleared and stained paratype of pertiviano, placed it in the genus Eleutherodaciijhis largely because a single T-shaped terminal phalange was found among the fragments of the skeletal preparation. Ill V Fig. 2.—Terminal phalanges of Phrtjnopus parkeri (Row A, KU 135306), P. pemvianus (Row B, figure at far left, KU 138928; other five phalanges of KU 138926), and P. cophites (Row C, KU 138909). Column to left (no roman numeral ) is of tliird finger. The five columns headed by roman numerals include terminal phalanges on toes I tlirough V. 6 OCCASIONAL PAPERS MUSEUM OF NATURAL HISTORY Collections made in southern Peru by William E.